**** BEGIN LOGGING AT Fri Feb 15 02:59:58 2013 Feb 15 05:53:45 Hurrian, how do you revert back to stock camera Feb 15 05:53:52 app Feb 15 07:04:10 Hurrian, check you log you give the answer before Feb 15 07:31:03 ~karma chameleon Feb 15 07:31:03 chameleon has karma of 1 Feb 15 07:31:17 pff, no boy george reference **** ENDING LOGGING AT Fri Feb 15 08:38:45 2013 **** BEGIN LOGGING AT Fri Feb 15 08:40:19 2013 Feb 15 08:50:53 kerio, merlin1991, DocScrutinizer05: just realized that back when talking about eudev I ought to tell you the current channel for it is #gentoo-udev Feb 15 08:51:27 so you can go there and ask the devs directly Feb 15 08:55:21 * ccxCZ ponders about changing the led in open state to display paging/card io/wlan io Feb 15 10:54:45 11:54:14 up 983 days, 21:08, 1 user, load average: 1.04, 1.07, 1.07 Feb 15 10:55:13 :) Feb 15 10:56:25 and seeing that Feb 15 10:56:31 the sysop next to me just went "reboot". Feb 15 10:56:53 teotwaki: that's a lot of unpatched bugs Feb 15 10:57:12 jacekowski: yeah, it's also completely off the grid, so no real issues there. Feb 15 10:57:28 and I think they use the non-reboot kernel replacement. Feb 15 10:57:35 that's somewhere around 2.6.20 kernel probably Feb 15 10:57:46 you can't replace kernel without reboots Feb 15 10:57:52 ksplices can only patch very limited things Feb 15 10:58:23 2.6.18 :) Feb 15 10:58:46 jacekowski: to be honest, if someone gets into the network, they don't need kernel exploits to root the machines... Feb 15 11:17:23 teotwaki: and that's also bad Feb 15 11:17:33 jacekowski: technically you can replace the kernel without rebooting Feb 15 11:17:40 /dev/kmem is writable by root Feb 15 11:17:52 you have to be very good, though :P Feb 15 11:19:32 kexec might not technically count as a "real" reboot either. (Not that it probably existed 983 days ago.) Feb 15 11:19:47 fizzie: ksplice doesn't use kexec, i think Feb 15 11:19:56 they just backport patches as modules, or some weird shit like that Feb 15 11:20:03 and only security patches Feb 15 11:20:55 Sure, but it didn't seem to be entirely clear what "the non-reboot kernel replacement" was. Feb 15 11:22:22 and yeah, kexec is definetely cheating Feb 15 11:23:30 suspend to disk is cheating harder Feb 15 11:24:58 kerio: never said it was good :) Feb 15 11:25:13 SpeedEvil: wait, suspend to disk and then boot with a different kernel? Feb 15 11:25:22 ...would that even work? Feb 15 11:26:04 kerio: but frankly, there's very little chance anyone could get through. There's an ssh that only listens on an admin-only hard wire, which needs to be connected manually, all other traffic is pure application traffic. Feb 15 11:26:34 teotwaki: have you checked all the CVEs that apply to the current kernel? Feb 15 11:26:49 kerio: if there's no way to reach it, does it matter? Feb 15 11:27:17 maybe there's a remote execution/priviledge escalation bug that's exploitable via ICMP packets Feb 15 11:27:23 or ARP requests Feb 15 11:27:42 Erhm, how should I put this. Feb 15 11:27:58 kerio: there's nothing else on that network. Feb 15 11:28:05 then what use is it? :o Feb 15 11:28:11 oh, no Feb 15 11:28:24 same kernel Feb 15 11:28:47 SpeedEvil: what happens if you use a different kernel? Feb 15 11:28:54 shit breaks horrendously? Feb 15 11:29:11 panic Feb 15 11:29:21 actually, isn't the kernel loaded in ram anyway? Feb 15 11:29:24 kerio: we run a couple hundred servers, please don't assume you know everything there is to know. Feb 15 11:30:05 wouldn't the original kernel get restored? Feb 15 11:30:14 kerio: our use cases may differ vastly from what "classical" production does. Feb 15 11:31:53 hm, according to hitup.org there's a dude who has 11 years of uptime on a mac Feb 15 11:32:41 as far as i know it's impossible Feb 15 11:33:19 jacekowski: perhaps he's running os x beta? Feb 15 11:33:33 even 10.0 would work, it's been released in 2001 Feb 15 11:33:35 iirc, there is something about how the uptime is stored Feb 15 11:33:39 Impossible? Feb 15 11:33:49 and clock will reset after 420 days or something Feb 15 11:34:08 Max Uptime:17 Years 6 Days 21 Hours 47 Minutes Feb 15 11:34:14 but it's a cluster, i don't know if it counts Feb 15 11:34:17 http://www.uptimes-project.org/hosts/view/150 Feb 15 11:34:46 uuuuu Feb 15 11:34:52 95.32 idle Feb 15 11:34:55 Total Uptime:86 Years 63 Days 9 Hours 47 Minutes Feb 15 11:34:59 sounds fake for a cluster Feb 15 11:35:01 wait, what? Feb 15 11:35:39 uptime * noofprocessors Feb 15 11:35:56 which is a pretty stupid metric Feb 15 11:36:14 http://www.uptimes-project.org/hosts/view/27 ok, this website is bullshit Feb 15 11:36:51 not if he has 4x 8 core xeon Feb 15 11:36:56 with HT Feb 15 11:37:03 that would be 64 visible cores Feb 15 11:37:16 hm, 4 years Feb 15 11:37:57 but yeah, whole uptime thing is easy to fake Feb 15 11:40:20 jacekowski: more importantly, where's the nuclear reactor to power the xeons? Feb 15 11:40:48 dont forget the cooling towers! :p Feb 15 11:41:04 Cool Towers? I use the moon as my heatsink, bitches. Feb 15 11:41:11 :P Feb 15 11:41:25 space is actually quite a crappy heatsink Feb 15 11:41:40 You say space, I say the moon. Feb 15 11:41:43 It's filled with cheese. Feb 15 11:41:46 Awesome heatsink. Feb 15 11:41:50 but then the cheese would melt! Feb 15 11:42:02 Duh Feb 15 11:42:04 Of course not. Feb 15 11:42:09 If cheese melted Feb 15 11:42:15 The sun would've taken care of that years ago. Feb 15 11:42:22 right, how stupid of me Feb 15 11:42:31 And we'd just have a ball of fondue floating around planet Earth. Feb 15 11:42:37 Hence, cheese doesn't melt. Feb 15 11:43:27 what if we *do* have a ball of fondue Feb 15 11:43:35 If we did Feb 15 11:43:40 and the moon landings were staged because you can't walk on fondue Feb 15 11:43:44 Then the mirrors on its surface wouldn't reflect properly. Feb 15 11:44:02 you put the mirrors *over* the fondue Feb 15 11:44:08 it helps to glue them in place Feb 15 11:44:11 yeah, but it'd drift away Feb 15 11:44:15 it's fondue dude Feb 15 11:44:21 the mirrors would sink Feb 15 11:44:39 transparent cheese Feb 15 11:45:29 anywayu Feb 15 11:45:40 983 days was nice to see :) Feb 15 11:45:47 "was"? D: Feb 15 11:45:47 Probably our record Feb 15 11:45:48 you monster Feb 15 11:46:01 and seeing that Feb 15 11:46:01 the sysop next to me just went "reboot". Feb 15 11:46:12 oh, he actually did it ._. Feb 15 11:46:24 we migrated the services from it last night Feb 15 11:46:45 so the machine will now be reinstalled/repurposed for backup duty. Feb 15 11:48:05 you seen iphh.net server moving video? Feb 15 11:48:12 * kerio did Feb 15 12:51:38 DocScrutinizer05: link please Feb 15 12:51:58 ask kerio, I lost it Feb 15 12:52:13 ask the chanlog Feb 15 12:52:14 just busy with mashiara, sorry Feb 15 12:52:44 there you go: http://www.youtube.com/watch?v=vQ5MA685ApE Feb 15 12:53:05 teotwaki: haha, i didn't realize it, the server had a 7-years uptime Feb 15 13:21:55 Pali: :-D good mail Feb 15 13:22:15 pastebin pls Feb 15 13:22:20 Pali: Niels is working on autobuilder to fix it, since err... a week Feb 15 13:22:34 DocScrutinizer05, sorry but from my point of view this seems like maemo is turned off Feb 15 13:23:00 maemo developers cannot do anything Feb 15 13:23:14 maemo testers also Feb 15 13:23:43 Pali: we (council, tech staff) are not happy with this either. Just there isn't that much we can (competence, access) or are allowed to (HiFo approval, access) do Feb 15 13:24:26 I (as cummunity developer) did not received any message Feb 15 13:24:28 Pali: I constantly pushed behind the scenes. A week ago Eero told us that X-Fade is now working on it to fix it Feb 15 13:25:26 DocScrutinizer05, at least somebody from board/council/... should write message to maemo developer mailinglist whats happening Feb 15 13:26:08 to give credit where credit is due, it probably been last week's meeting between Nokia, Nemein and board that finally made things move Feb 15 13:26:16 because silence for community developers means it is going killed Feb 15 13:26:54 Pali: please join tonight's maemo council meeting and bring it up Feb 15 13:27:26 DocScrutinizer05, time & channel Feb 15 13:27:51 Pali: #maemo-meeting every friday, 18:00 UTC Feb 15 13:27:52 my plate is more than filled with critical migration management, so maybe one of the other council guys can pick it up and send a mail to ML Feb 15 13:28:07 kerio, ok Feb 15 13:29:45 technical sidenote: it seems that root culprit is account/auth management located on garage Feb 15 13:30:31 http://roundup.fourecks.de/maemo/issue3 Feb 15 13:32:05 it is really good to know that things are moving in the right direction ! :) Feb 15 13:32:13 thanks to all involved BTW ! :) Feb 15 13:33:10 I'm using a temporary repository in the meantime, but considering that most N900 users probably don't read TMO or are not as advanced as to add a new repository Feb 15 13:33:28 most probably think I didn't update modRana since November :) Feb 15 13:33:46 so I'm relly looking forward to a working autobuilder :) Feb 15 13:44:02 Pali: is it possible to add "keys" or whatever they're called to a hal device entry? Feb 15 13:44:44 kerio, I'm sure that it is possible to update values (some only as root) Feb 15 13:44:44 better question: is it possible to add keys to a hal device entry dynamically? Feb 15 13:44:55 or do they have to be specified somewhere? Feb 15 13:45:03 so I think it should be possible to add also Feb 15 13:45:51 kerio, call: libhal_device_set_property_string Feb 15 13:45:59 (or libhal_device_set_property_int) Feb 15 13:46:10 or other libhal_device_set_property_* function Feb 15 13:46:34 i was thinking that hald-addon-bme should be the one providing time-to-empty and time-to-empty-if-idle Feb 15 13:46:46 and percentage Feb 15 13:46:51 so the battery applet becomes much, much simpler Feb 15 13:47:25 especially because hald-addon-bme can decide to provide those values straight from bq27k, when appropriate Feb 15 13:55:37 yes, that's the way to go Feb 15 14:08:41 it's quite disappointing that HiFo (Robert) _can_ answer mail from Nemein within 2h, while mail from council - however carefully and politely worded and of whatever high urgency and relevance - gets ignored same time Feb 15 14:12:40 I have a *very* hard time trying to apply the good old principle of "never attribute to malice..." here Feb 15 14:31:58 kerio, DocScrutinizer05: n900 timer resolution seems to really be 50ms, verified using gettimeofday. so I assume kernel HZ is 200 Feb 15 14:32:27 daaamn Feb 15 14:32:31 that sucks :s Feb 15 14:32:49 but I found another way to get a high res tick. I'm using an audio recording cannel with 4000 hz sampling rate and count samples ;) Feb 15 14:34:01 ^^ but that only works flawlessly as I need the time interval between events (spikes) within the audiostream as well Feb 15 14:34:19 ohohoh, better wipe PA out of your picture then. direct alsa card may work, PA most certainly won't Feb 15 14:34:30 ? Feb 15 14:34:49 NB that PA offers a fake ALSA card to apps Feb 15 14:35:36 topro: instead of an audio recording, use alsa's snd_timer_read() Feb 15 14:35:47 aiui it should do the same thing for you Feb 15 14:36:12 yeah, but on a *real* alsa device, not on PA's alsa compatibility layer fake Feb 15 14:36:39 ah, now I see. thats an interesting option. but as you say it works fine for me just counting samples. cpu load is ~3% Feb 15 14:37:14 as my code has to iterate through theses audio samples anyway to detect spikes there is nearly no overhead Feb 15 14:37:31 but jitter Feb 15 14:37:48 if this is a normal audio stream that's managed by PA Feb 15 14:37:58 jitter because I get those samples in packets you mean? Feb 15 14:38:03 it is NOT realtime Feb 15 14:38:50 well actually I don't even know what backend I use, I'm usinf QAudioInput Feb 15 14:39:06 it's however exactly the right thing to analyze audio, since you basically got implicit timestamps on each sample, for non-realtime deferred processing Feb 15 14:39:41 thats exactly what I am doing, realtime doesn't matter in the sub-second scale Feb 15 14:39:52 ^^ Feb 15 14:39:59 for my current task that is Feb 15 14:39:59 then your approach is basically the only correct one Feb 15 14:41:05 and PA doesn't matter either in this case Feb 15 14:41:11 if i would need realtime, snd_timer_read() would be my friend using plain alsa? Feb 15 14:41:21 yes Feb 15 14:41:35 could you please explain PA to me? Feb 15 14:42:23 app -> PA-fake-ALSA -> PA-soundserver -> real (ALSA) hw audiocard Feb 15 14:42:47 well actually what does "PA" stand for? Feb 15 14:42:57 PukeAudio Feb 15 14:43:04 ah, sorry Feb 15 14:43:10 err Polypaudio, nah PulseAudio Feb 15 14:43:24 ~pa Feb 15 14:43:25 it has been said that pa is PAY ATTENTION!!!!! Feb 15 14:43:29 ugh Feb 15 14:43:39 gg Feb 15 14:44:05 another poettering abomination Feb 15 14:44:10 so there is a real alsa hw on n900 I could use to get timer ticks? Feb 15 14:44:35 yes, PA sits on top of a real ALSA kernel implementation Feb 15 14:44:57 the kernel soundcard hw drivers are ALSA Feb 15 14:45:13 so what would I have to take care of to get ther right backend? Feb 15 14:45:28 good question Feb 15 14:46:14 I guess as long as PA is running, access to real ALSA *stack* is hidden from apps. What is available are the alsa soundcard stubs in /dev/sound Feb 15 14:46:38 though might be locked by PA having them opened Feb 15 14:46:56 os this just to get timer? Feb 15 14:47:02 a timer Feb 15 14:47:13 or for use with sound Feb 15 14:47:44 to get a high res timer Feb 15 14:47:53 I dunno. For high resolution timers there are other better alternatives than soundcard Feb 15 14:48:03 on n900? Feb 15 14:48:17 i thought soundcards were really good at timing Feb 15 14:48:48 you could e.g. exploit one of the PWM counters Feb 15 14:48:57 you don't use them as clocks if you've got clocks Feb 15 14:49:18 I thought there was a high res timer anyway Feb 15 14:49:25 is it not eszpe? Feb 15 14:49:28 exposed Feb 15 14:49:52 it should also be available, yes. AFAIK Feb 15 14:50:20 FIQ is another possible approach, depending on what you want to do Feb 15 14:51:45 ok, i see there are some possibilities. for the moment I need the time between two spikes in audio stream, so I think I will go with counting samples. If one day I would need another real high-res timer I'll evaluate some of your proposals ;) Feb 15 15:50:33 topro; i would assume tickless kernel Feb 15 15:50:52 topro: for timing between two audio spikes, counting samples is ablutely the way to go Feb 15 15:50:55 The only thing you'd maybe want a timer for, is to determine whether actual samplerate is the requested 44100 or 44099 Feb 15 15:50:58 do that over longer period and system timer granularity doesn't matter as much Feb 15 15:51:54 ShadowJK: thanks Feb 15 16:09:27 FIQ is another possible approach, depending on what you want to do Feb 15 16:09:28 what? Feb 15 16:09:33 oh wait Feb 15 16:09:36 it was not pointed to me Feb 15 16:09:37 nvm Feb 15 16:10:34 your fault ;-P Feb 15 16:10:55 meh :P Feb 15 16:11:10 I've used this nick for way too long to be interested in changing it Feb 15 16:11:19 np Feb 15 16:11:26 sorry for highlight Feb 15 16:13:12 yay, it's really funny how one can spend the whole day, week by week, with just managing migration issues. I just suggested hidden primary for our DNS issues, let's see if that will fly Feb 15 16:14:01 Nemein/Eero is not averse and will support us Feb 15 16:14:24 even donate a location and IP for the NS Feb 15 16:25:11 http://www.h-online.com/open/news/item/Vivaldi-tablet-in-harmony-with-new-vendor-1804422.html Feb 15 16:27:12 yay, if our migration business continues developing the way it does, I start to wonder if we ever need to do the final big cut Feb 15 16:28:03 but it seems Nokia will insist in that big cut, to get rid of responsibility for maemo.org Feb 15 16:30:58 topro: may I ask what nifty stuff you're building? Feb 15 16:31:36 acoustic ping? Feb 15 17:22:34 http://www.bbc.co.uk/news/world-europe-21468116 boom Feb 15 17:23:03 wow (not directly on topic) - meteor Feb 15 17:24:34 oooh. altimeter chips in phones could as a distributed network do energy analysis of such blasts Feb 15 17:25:31 SpeedEvil: (meteor) yeah, 28000km distance, allegedly. But when, exactly? Feb 15 17:25:48 today Feb 15 17:25:56 12hish ago Feb 15 17:26:08 900 injured Feb 15 17:26:09 oooh Feb 15 17:26:27 nah, not the tiny meteorites in .ru Feb 15 17:26:37 the "big" one Feb 15 17:26:47 tonight Feb 15 17:27:11 (65m big iirc) Feb 15 17:27:44 allegedly passing by in only 28000km distance Feb 15 17:28:06 not very tiny, probably a few m in diameter Feb 15 17:28:17 it theoretically might hit satellites Feb 15 17:29:03 or earth if they did an oopsie in their calculations Feb 15 17:29:11 easy to compute. Feb 15 17:29:45 yeah, but 28000km is *really* tight Feb 15 17:31:07 and I honestly can't believe the .ru incident is by pure chance Feb 15 17:31:14 not as tight as 0 Feb 15 17:31:53 anyway, you do know when it's supposed to be most close to us, aka "pass by"? Feb 15 17:33:03 tonight, 8 pm ish for you Feb 15 17:33:11 binoculars Feb 15 17:33:21 and good sky needed Feb 15 17:33:21 SpeedEvil: *if* they knew it will hit somewhere, but can't exactly calculate *where*, don't you think they'd tell exactly this: it will pass by very close Feb 15 17:33:57 I meant zero for the one that hit today Feb 15 17:34:52 also, it was observed closely yesterday. orbit is very well known Feb 15 17:35:48 they mustn't tell "it will impact somewhere in $arbitrary-huge-area-with-millions-of-people", since that would for _sure_ cause a high number of fatalities even when the meteor would miss earth Feb 15 17:37:34 they won't lie about the timing though, probably Feb 15 17:37:40 there is no they Feb 15 17:37:49 yeah, sure Feb 15 17:38:02 amateur s also picked it up Feb 15 17:39:10 I'm pretty sure they did. But can they do *exact* measurements of orbit parameters and then do the calculations to a precision better than 28000km error? Feb 15 17:39:48 I'd guess velocity measurement is a quite demanding process Feb 15 17:41:19 you measure it against the fixed star background Feb 15 17:41:40 that's tangential velocity relative to us Feb 15 17:41:58 indeed Feb 15 17:42:09 axial velocity directly towards us is hard to measure this way Feb 15 17:42:39 correlating with older images lets you derive an orbit very closely Feb 15 17:42:50 there's basically only radar and size-changes Feb 15 17:43:27 yeah, if complete orbit parameters are known, you can use pure math Feb 15 17:44:24 unless those parameters changed by events that happened while the object not been in our vicinity Feb 15 17:50:27 collecting a bit of dark dust so albedo changes is enough to change orbit parameters enough for an error of >28000km, give the speed of earth on its orbit around sun Feb 15 17:50:32 given* Feb 15 17:53:11 the radial velocity can be closely estimated in most cases from the tangential velocity evolution and orbital mechanics Feb 15 18:01:18 DocScrutinizer05: could you pastebin the output of `dpkg --get-selections`? i'm still trying to figure out wtf is wrong with my Conversations, and maybe it's caused by something that i deleted Feb 15 18:01:43 no, since now is meeting time Feb 15 18:01:55 but it'll take a second! :( Feb 15 19:07:49 SpeedEvil; someone on radio said the orbital params of the russia meteors don't match the one being tracked Feb 15 19:37:51 I'm absolutely sure, that today's meteor shower is not corellated with any asteroid, but it's omen of DocScrutinizer05 leaving administrator position, fools. Feb 15 19:38:01 OTOH, quick question Feb 15 19:38:45 what si the most lightweight solution to manipulate content from live webpage on n900 itself, aka equivalent of firefox'ish "live http headers"? Feb 15 19:39:06 reason is trivial - me wanna to grab some swf file from some site that obscure it Feb 15 19:39:27 could use desktop for that, but I'm pretty sure N900 should be able to :) Feb 15 19:41:20 and, question 2 - what is, actually, the most lightweight from available solutions, to create a http sever on N900 - for localhost use only, in fact to "cheat" swf files, that have hardcoded requiments to run from certain domain Feb 15 19:41:33 (so i "create" that domain and run it from localhost Feb 15 19:44:37 Estel_: i'd say busybox-power's httpd :) Feb 15 19:44:54 i'm not sure of how lightweight it is, though, compared to lighttpd or nginx Feb 15 20:19:59 i've used busybox httpd on a 256MB x86 (266 MHz Geode GX1)... seems to work ok, doesn't overwhelm the system. it works with cgis too... but unsure of security :o no ssi though... Feb 15 20:21:20 the n900 is faster than the gx1 at least... Feb 15 20:22:42 eccerr0r: it would be strictly local-only, if i understood Estel_ correctly Feb 15 20:22:53 also, afaik busybox httpd is used by dd-wrt Feb 15 20:24:03 n900=5x geode gx1@266 Feb 15 20:25:43 as long as it's configured correctly... httpd needs to be run as root to bind to port 80, which it will happily do... Feb 15 20:41:12 eccerr0r, kerio, thanks a lot Feb 15 20:41:23 lol, never thought about busybox's httpd Feb 15 20:41:25 btw Feb 15 20:41:41 any ideas about first part of question, i.e. dissecting http headers? Feb 15 20:42:11 to get GET calls to actual content that I want download, that is, otherwise, obscured? Feb 15 20:42:48 tcpdump port 80 -s 0 -o outfile Feb 15 20:42:55 open with wireshark Feb 15 20:43:11 * ShadowJK forgets if -s 0 is infinite capture length, and if -o is the right option Feb 15 20:47:41 Estel_: not sure about your use case, but `python -m SimpleHTTPServer' is a pretty simple httpd Feb 15 21:00:03 gregoa: blah Feb 15 21:01:03 kerio: I'd prefer "plackup -MPlack::App::Directory -e'Plack::App::Directory->to_app'" but I don't think libplack-perl is in maemo :) Feb 15 21:07:42 gregoa: twistd web --path . Feb 15 21:08:30 kerio: nice, thanks Feb 15 21:09:00 twistd is the multi-purpose Twisted daemonizer/launcher Feb 15 21:11:49 hey guys, you know i usually have "hide join/part/nick", just temporarily enabled it for council meeting. Why nobody of you pinging chanops about annoyance like <-- thomasvs has left this server (Excess Flood). Feb 15 21:12:11 meh Feb 15 21:13:10 /mode +b $a:thomasvs$##fix_your_connection Feb 15 21:13:28 your wish is my command Feb 15 21:13:31 if we want to get all fancy Feb 15 21:13:46 although... Feb 15 21:13:54 if that's what you want me to do Feb 15 21:14:02 it seems to have stopped Feb 15 21:14:25 DocScrutinizer05: you're a good op, your intervention fixed it :3 Feb 15 21:14:47 dammit! Feb 15 21:15:11 DocScrutinizer05: he doesn't seem to join when already identified Feb 15 21:18:08 hm? Feb 15 21:18:14 that thomasvs dude Feb 15 21:18:25 he answered three times to a ctcp version Feb 15 21:18:27 I've seen him changing host, yeah Feb 15 21:18:33 DocScrutinizer05, as for myself, I also have join/part hidden :P Feb 15 21:23:08 hmm, now? Feb 15 21:24:20 DocScrutinizer05: /mode #maemo +b $x:thomasvs*$##fix_your_connection Feb 15 21:24:23 i'm fairly certain of that Feb 15 21:24:41 (redirect to ##fix_your_connection which is /the/ channel for this) Feb 15 21:26:13 I use join/part hiding in pidgin, set to hide if user was idle for >10 min Feb 15 21:33:02 Thank you Feb 15 21:33:17 I doubt this been a wellformed ban Feb 15 21:33:42 looks kinda strange to me Feb 15 21:34:41 I'm just too lazy to check that weird syntax Feb 15 21:34:56 and semantics of x etc Feb 15 21:36:07 freenode extensions Feb 15 21:36:31 I heard of them but don't know them by heart Feb 15 21:37:24 DocScrutinizer05: it worked Feb 15 21:37:30 and too lazy to find the right webpage Feb 15 21:37:36 no, wait Feb 15 21:37:45 ...he was already there, lol Feb 15 21:38:02 ban doesn't kick Feb 15 21:38:09 yeah, i figured Feb 15 21:38:26 but noooow Feb 15 21:38:51 awesome Feb 15 21:38:59 he joined ##fix_your_connection :D Feb 15 21:39:13 but he was already redirected there by another channel, i think Feb 15 21:41:00 lol, myrtti is chanop there, why does this sound familliar to me? Feb 15 21:41:11 idk, why? Feb 15 21:41:14 also deop Feb 15 21:41:22 meh Feb 15 21:42:45 * DocScrutinizer05 shakes fist against smokedetector, rubs his ears and takes a breath from cig Feb 15 21:43:41 also buy an ecig Feb 15 21:43:50 meh have that Feb 15 21:43:59 ...do they trip smoke detectors up? Feb 15 21:44:04 *off Feb 15 21:44:18 sure they do, when you use them close to it Feb 15 21:44:29 but it's just vapour! Feb 15 21:44:31 but that "smoke" vanishes way faster Feb 15 21:45:00 so no building up of critical concentrations during council meetings Feb 15 21:45:28 :D Feb 15 21:45:28 next time I'll open the window before meeting stats Feb 15 21:45:31 starts Feb 15 21:45:46 why do you have a smoke detector anyway Feb 15 21:46:17 because I am declared dead by firebrigades, police, and paramedics Feb 15 21:46:30 I survived a fire in my flat Feb 15 21:46:43 don't want to try my luck again Feb 15 21:47:03 damn Feb 15 21:47:07 electric fire? Feb 15 21:47:25 probably dirt cheap desk vacuum cleaner Feb 15 21:47:36 overcharging NiMH Feb 15 21:47:50 hmm... chemical fire then? Feb 15 21:48:06 errr Feb 15 21:48:10 ionization smoke detectors will trigger on any particles/vapour etc. There the most sensitive type of detector. Feb 15 21:48:16 outch .. told ya, dont use the 787 cells Feb 15 21:48:29 anyway, i don't smoke and i don't clean my room Feb 15 21:48:33 so i hope i'm safe Feb 15 21:48:35 lol Feb 15 21:48:39 the AAA cells got hot, started plastic of vacuum to burn, and that ignited my shelf with books Feb 15 21:48:53 D: Feb 15 21:48:57 which books did you lose? Feb 15 21:49:02 all Feb 15 21:49:04 can you sue the producer for compensation? Feb 15 21:49:15 for sure not Feb 15 21:49:53 the true reason wasn't to find anymore in that ashes Feb 15 21:50:06 * cehteh buyed some ammo boxes for charging LiPo's Feb 15 21:50:13 but I know what been there that could've started a fire Feb 15 21:50:29 cehteh: isn't that... overdoing it? Feb 15 21:50:48 kerio: big ones for a hexacopter Feb 15 21:50:59 hm, then maybe it's more justified Feb 15 21:51:04 hehe, probably the 100 rounds 9mm ammo saved my life, when they went off in the fire and awakened me Feb 15 21:51:23 DocScrutinizer05: this made me realize, i don't know where (and if) the fire extinguisher is Feb 15 21:51:38 I *know* where mine is Feb 15 21:51:41